August 11th, 2016
August 11th, 2016
Inspired by Elon Musk’s recent post on Tesla's Master Plan, I felt it appropriate to try and map out what Kepler will look like over the next few years. But before I talk about our master plan, I think everyone needs a solid understanding of the why. That is why we’re building Kepler.
We get up every day because we know we’re building for a better future. We’re enabling a space economy that will allow for the continued advancement of a society that continues to stretch the carrying capacity of the Earth. We’re not alone in this endeavor; new companies are created every day, addressing the need for launch capacity, developing in-space resources, or democratizing space access. This is even more evident when you consider the number of satellites launched each year, which we (and many others) think will look something like this:
Much like how mobile networks are now a necessary part of our day-to-day lives, this coming boom in space access will create the same necessity off our planet. So we’ve set out to build the cellular infrastructure in space. While a noble goal, we still needed to make Kepler a viable business in the short term. To that end, we’ve mapped out a master plan that’s pretty simple and we think it will get us there. Here’s what it looks like:
Step 1 - Build a satellite platform 100x cheaper
At first glance this might seem like an enormous feat but when we take a deeper dive in the satellite industry it doesn’t seem all that audacious. The current satellite cost structures are centred around mainframe infrastructure, and waterfall development. If we take a few lessons from the PC and software industries, an agile development philosophy should work to realize these cost savings.
Companies like Planet & Spire have already realized these cost savings in satellite development and are changing the earth observation, AIS, and GPS-RO markets. We’re following in their paths by bringing it to telecom. We believe one of the major hurdles to making this realizable is developing a high bandwidth, power efficient radio, and we think we have that figured out.
Step 2 - Sell a niche service to early adopters
At this point, we'll have proven we can build tech that works at our target price, but we won't yet be ready to scale the business because we won't have proved that our market is ready to pay for it. So we’re adopting an incremental approach where with a small amount of capital and something much smaller than our full constellation we can have early customers paying (real cash – not LOIs) for our service.
We’ve taken this approach because of all the data we gathered interviewing founders and early employees of some of the most substantial satellite projects including Teledesic, Orbcomm, and Iridium (you can't expect to run a successful satcom company without asking these guys why they failed!). While they all provided a great deal of detail on why they failed – maybe that’s for another blog post – they generally shared one common theme for failure. That is thinking that “if you build it, they will come”. So they set out to deploy entire constellation all in one shot, without paying enough attention to costs and market demand. Over time their cost vs. market demand chart looked something like this:
This is something we'd like to avoid, which is why we've built this incremental approach
Step 3 - Use the money to fund mass production
By step 3 we’ve now proved a lot of our hypotheses around low cost satellites and being able to sell a service. From this point on we’re comfortable scaling our satellite production to a number greater than the first phase but less than our final constellation. Yes, the number of satellites is intentionally vague because there are a lot of variables at play.
Step 4 - Sell a cheaper telecom service
Once we’re at step 4 we’ve probably launched our full constellation and are providing the telecom service we intended with Kepler. Though it doesn’t end there for us. We will need to keep innovating to hit internal price reduction targets. This will allow us to explore other ways to use this small satellite constellation for different applications.
That’s our master plan. We’ll come back to this in a few years to see how accurate it was.